Neil Gandal, Matan Yonas, Michal Feldman, Ady Pauzner, Avraham Tabbach 25 June 2020
The COVID-19 pandemic continues to spread around the world. Estimates indicate that together with the large numbers of deaths (more than 450,000 as of 22 June), there will also be significant economic consequences (Baldwin and Weder di Mauro 2020a, 2020b). Scientists from all disciplines are targeting their efforts to understand the phenomenon and offer ways to combat it. Our research (Gandal et al. 2020) aims at highlighting the role of long-term care facilities in spreading the virus to the most vulnerable population (i.e. old people). This group has suffered the most deaths and has also overwhelmed local health systems.
Similar to Bayer and Kuhn (2020), our starting point is the large variations in COVID-19 death rates across countries, and in particular across European countries, where the pandemic’s first wave seems to be over. (see Figure 1).
Figure 1 Covid-19 deaths up to mid-May 2020 (per million)
How can these differences be explained and what are the policy implications? One might suggest that the death rate is higher in countries with older populationa. Another possibility is that a higher death rate is linked to countries having poorer health care systems. For example, Ciminelli and Garcia-Mandicó (2020b), show that within the area of the epidemic epicentre in Italy, the mortality rate was up to 50% higher in municipalities far from an intensive care unit (ICU), a sign that congestion of the emergency care system may have prevented critical patients from being treated on time. Bayer and Kuhn (2020) attribute the differences in fatality rates to intergenerational interactions. They argue that countries with more intergenerational interactions (where the young live with the old) have higher case fatality rates.1
Our study focuses on the effects of the living arrangements of the old population – specifically, living in long-term care facilities – on COVID-19 deaths. One critical issue that has received a great deal of media coverage is the high rates of the deaths worldwide from COVID-19 that have occurred among residents of long-term care institutions. Euronews reported that deaths among such residents could account for more than 50% of all COVID-19 deaths in Europe.2 According to an article in The Guardian, data from the Kaiser Family Foundation indicates that deaths among long-term care residents account for more than 50% of all deaths attributed to COVID-19 in 14 states in the US. The same article also notes that in the state of New Hampshire, 72% of COVID-19 deaths occurred among those living in long-term care settings.3 Overall, according to the New York Times,4 more than one third of the deaths in the US from COVID-19 have occurred among long-term care residents. The U.S. Center for Disease Control and Prevention (CDC) has formally stated that, generally, people aged 65 and older – and in particular, “people who live in a nursing home or long-term care facility” – are at high-risk of severe illness from COVID-19.5
Taken together, there is clear evidence that a large percentage of the deaths from COVID-19 have occurred among residents of long-term care facilities. Importantly, there are two possible reasons for this phenomenon.
- First, the structural features of such settings – a communal living area, multiple residents in a room, care provided by multiple caregivers to multiple care recipients, etc. – lead to excessive death.
- Alternatively, it is possible that individuals in these facilities are in poorer health than those living elsewhere, and that these individuals would have died even if they had not been in these facilities.
Whether living in a long-term care facility causes death or is just associated with other risk factors has very different policy implications.
We examine these competing explanations by studying the association between long-term care beds per capita in a country and COVID-19 deaths per capita. We employ country-level data from Europe, since Europe is the only region in the world in which there are comprehensive (and standardised) data on long-term care bed capacity at the country level.6 Figure 2 illustrates the association between the two measures.
Figure 2 Relationship between COVID-19 deaths per capita and long-term care beds per capita, 32 countries (logarithmic scale)
Using these data, we develop a structural model and estimate the relevant ‘reduced-form’ equation which studies how deaths per capita in a country is explained by the number of long-term care beds per capita, controlling for other important drivers of the death rate (the percentage of the population aged 75 and over, population density, and hospital beds per capita).
We find that these four exogenous factors are all statistically significant and explain nearly 70% of the variation in the death rates among European countries. Importantly, we find that, controlling for the other variables, there is a positive and statistically significant association between the number of long-term care beds per capita and COVID-19 mortality rates per capita for countries in Europe.7 Our analysis thus supports the first competing hypothesis, namely that the structural features of long-term care facilities may lead to death.
These findings are, of course, preliminary, and we are planning to extend them to at regional levels of the different countries once we obtain the data.8 Nevertheless, even at this early stage, our findings raise important policy implications. In particular, they suggest that efforts should be geared to protecting older adults living in long-term care settings. Policymakers might even consider alternative dwelling options during the epidemic period, such as encouraging residents to live with their families whenever possible.
Baldwin, R and B di Mauro (2020a), Economics in the Time of COVID-19, a VoxEU.org Book, CEPR Press.
Baldwin, R., and di Mauro, B. (2020b), Mitigating the COVID Economic Crisis: Act Fast and Do Whatever It Takes, a VoxEU.org Book, CEPR Press.
Ciminelli, G and S Garcia-Mandicó (2020a), “COVID-19 in Italy: An analysis of death registry data”, VoxEU.org, 22 April.
Ciminelli, G and S Garcia-Mandicó (2020b), “COVID-19 in Italy: An analysis of death registry data, Part II”, VoxEU.org, 19 May.
Bayer, C and M Kuhn (2020) , “Intergenerational ties and case fatality rates: A cross-country analysis” VoxEU.org, 20 March.
Comas-Herrera, A, J Zalakaín, C Litwin, A T Hsu, N Lane and J L Fernández (2020), “Mortality associated with COVID-19 outbreaks in care homes: Early international evidence”, LTCCOVID.org, International Long-Term Care Policy Network, CPEC-LSE, 3 May.
Gandal, N, M Yonas, M Feldman, A Pauzner and A Tabbach (2020), “Long-Term Care Facilities as a Risk Factor for Death Due to COVID-19”, CEPR Discussion Paper 14844.
Sá, F (2020), “Socioeconomics determinants of COVID-19 infections and mortality: evidence from England and Wales”, COVID Economics: Vetted and Real-Time Papers 22, 26 May.
1 Other explanations have been offered. Ciminelli and Garcia-Mandicó (2020), for example, examine the effectiveness of different lockdown policies in Italy using death registry data. They find that shutting down non-essential services reduced mortality, while closing factories did not. In another study, Sá (2020) uses data on infections and mortality for small regions in England and Wales to study the association between socioeconomic factors and COVID-19. They find that areas with large households and areas with greater use of public transport have higher infection rates. They stress the importance of reducing the risk of infection on public transportation.
2 See here; see also the report by Comas-Herrera et. al., 2020.
5 https://www.cdc.gov/coronavirus/2019-ncov/need-extra-precautions/groups-at-higher-risk.html. See also Comas-Herrera et al. (2020) who document mortality associated with COVID-19 in the case of long-term care homes in different countries.
6 Data on long-term care beds per capita are available for all but a small number of European countries.
7 Long-term care beds alone explain 28% of the variation of the death rate. This results also helps to explain some of the wide and unexpected variation in the death rates from COVID-19 among European countries.
8 We also repeat this exercise for 21 regions within Italy and obtain similar results.